Animals will be made experimentally diabetic or hypertensive so as to produce diabetic retinopathy (DR) or hypertensive retinopathy (HR). Both are known to involve the function of the blood-retinal barrier (BRB) located in the cells of the retinal blood vessels and in the retinal pigment epithelium (RPE). Our aim is to investigate corresponding morphologic changes and to compare them with the morphology of the normal BRB. We will focus upon: 1. The content of actin within the cells of the BRB (overall content, distribution and relation to junctions within endothelial cells, pericytes and RPE cells). 2. Permeability of the barrier, particularly of the RPE, to the tracers fluorescein and horseradish peroxidase (localization and pathway of leakage, extent of barrier repair). 3. Possibility to influence blood vessel function and morphology by using pharmacologic agents (cytochalasins B and D) known to interfere with actin polymerization. Methods used are clinical observation, light microscopy of tissue cross sectios and of trypsin digest preparation; fluorescence microscopy of freeze-dried and of fixed tissues; electron microscopy of tissues containing electron microscopic tracers and of tissues prepared with myosin subfragment-1 for the demonstration of actin filaments. Information will be derived from the disciplines of clinical ophthalmology, ophthalmic anatomy and pathology, pharmacology and biochemistry and will be integrated to achieve a better understanding of the BRB in health and retinal vascular disease, notably DR which has become the most frequent cause of adult onset blindness.